In our partial replication of John Ridley Stroop’s experiment, these conflicting stimuli were investigated, and the outcome was later known as the Stroop Effect. This experiment compares the reaction time between reading out colors and identifying the ink in which they are written on a card. For the latter, the words act as mediators which come in the way of the recognition of colors the words are printed in. In other words, it requires identifying something other than what is indicated by the semantic meaning of the word. With two conflicting stimuli, comparing the reaction times for both demonstrates which is dominant over the other. As experimenters, we were given a card with a table of colors written in ink that did not correspond to the word. We then took a randomly selected sample of subjects from a controlled age group to run the experiment on.
With respect to the research question, I predicted that the conflicting stimuli will alter one’s attention, and the stimulus which has been reinforced and preconditioned in the past will probably yield a more prominent response. Therefore, in our Stroop Test, I hypothesized that the time required to identify the colors would be significantly greater than that taken to read the words. This is because the age group we selected for our sample consisted primarily of students studying in an English medium, and had fully developed rudimentary reading and writing skills.
Method
Design
The type of design for this was a controlled laboratory experiment, where one variable was changed (independent) in order to observe variance in another (dependent), while all others were kept constant. The independent variable was the stimulus, or the color card. We, as experimenters changed the response of our subject (word vs. color recognition). This resulted in a change of reaction time for the participants; the dependent variable.
Prior to conducting the experiment, all subjects signed a consent form in order to ensure their willingness to participate. Furthermore, they were briefed and offered the chance to ask any questions or even choose not to participate at all. During the experiment, we took great heed of the fact that the participants were not hurt in any way. After collecting all raw data, the subjects were informed of the goal and purpose of the experiment, and assured that no mental or physical harm had been done to them.
Participants
The total sample group consisted of eighteen participants, ages ranging from 14 to 15 years, with the mean or average being 15. There were eight females (two aged 14 and six aged 15) and ten males (two aged 14 and 8 aged 15). All subjects are from Year 10 at Dubai International Academy, and were selected at random in order to avoid any cultural or gender biases. For the conduction of the experiment, each experimenter (or pair of experimenters) was allocated two subjects, which were again chosen at random. This group remained constant for both trials of the experiment.
Apparatus/Materials
For this experiment, we required a signed consent form from each participant, a standardized briefing and debriefing sheet outlining exactly what the subjects needed to be told, a Stroop color card which was constant for each group and trial, and finally a stopwatch.
Procedure
This experiment was conducted under controlled laboratory conditions, with each group seated separately and given individual apparatus to use. Once all participants and experimenters were seated, the briefing began. Both the nature of the experiment along with the participant’s rights were outlined as clearly as possible, after which the subject was allowed any questions before commencing with the experiment.
Each group consisted of two participants, and either one or two experimenters. The experiment was conducted on each subject one at a time.
First, the experimenter noted down basic information for each participant, which would not be disclosed to any outside party. The gender, age and nationality of both participants were recorded.
The first test was to read the words on the card provided; either participant may go first. Their time taken to read the words is measured and recorded by the experimenter. Next, the same subject is asked to identify (out loud) the color each word is printed in, and this is once again timed by the experimenter. Moving on to the second participant, the same two tests are repeated in exactly the same manner and the data is now recorded for the other subject.
After having taken down two pieces of data for each participant, the entire experiment is repeated; test 1 for participant 1 onwards. Once both trials are done and all data has been recorded, the experimenter must debrief the participants based on the standardized sheet provided.
Results
The raw data for each participant is recorded in tabular form in the appendices. In order to make comparisons, the data was divided into two groups for each test: 14 year olds and 15 year olds.
The mean reaction time represents the average time taken for each age group to recognize either the words or their color, and provides a central point towards which the rest of the data tends. Standard deviation measures the dispersion of reaction times measured, or how different or far apart the data is. The table below portrays the approximate mean and standard deviation values for both age groups.
The table above shows a much higher reaction time for recognizing colors than for reading words, and this applies for both age groups. The standard deviation or dispersion of data was also far greater for color recognition, however for the data set of 15 year olds, it was significantly greater. This is most likely due to the fact that there was a larger number of 15 year olds in the sample, therefore there were more reading taken which increased the total distribution of the data. Overall, the fact that reaction time for color stimulus was significantly greater than that for the words agrees with my prediction before the experiment was conducted. This validates the data collected of the total sample as a whole.
However, in comparing the reaction times between 14 and 15 year olds, no conclusions can be drawn. One would expect an older student to be able to read the words or identify colors faster, however the information above demonstrates exactly the opposite. The reason for this is probably the sample size is much greater for 15 year olds therefore the data is more spread out. Perhaps if both age groups had equal numbers of participants, this comparison would hold more weight. Thus, for this particular experiment the data does not hold validity for comparison between ages.
Discussion
These results show a definite causal relationship between the words and their colors. When simply asked to read the words, all participants had a lower reaction time than when asked to identify the ink the words are printed in. For participants of age 14, reading the words took approximately 51% less time than identifying their colors, while for the 15 year olds it took up to 60% less time. The reason for this is probably that the data set for 15 year olds was larger, and as results above show, there was a greater dispersion for this set. My hypothesis stated that the time taken to identify colors when two conflicting stimuli were presented would be greater, and this is evidently supported by the results above. The different groups were kept under almost identical conditions however one thing that did vary was the experimenter. This brings out the possibility of experimenter bias, wherein during a briefing the experimenter could have accidentally revealed his/her hypothesis to the subjects, thereby altering the results.
In general, we found that reading words took less time than identifying colors, because the latter involves conflicting stimuli. The word acts as the mediator, and the color is actual stimulus. The reaction time is the response, which gets hindered by the text (word) and alters the participant’s classification of the color. The reason for this is that a student who has learned to read and write English would automatically read the word before looking at its font color. However, if the participant was someone who didn’t know how to read or write English, for example a kindergarten student, they would almost certainly be able to identify the colors faster. This is because the word no longer acts as an interfering stimulus, because no associations can be made to it. Going back to Schumann and Muller’s idea of links between two factors, we can see how these associations are formed through learned experiences. In the Stroop test, it is almost as if an association is being broken in order to distinguish the color without letting the word come in the way, which is why it takes so much longer. Kline’s ideas were similar however he proved this using a positive relation between associations. For example, if the words were printed in the same color as the text itself, this would probably speed up the process of reading them. This is because we associate the color with what is written, and vice versa. With regard to the Stroop effect, various other factors can be tested in order to see what influences reaction time and what doesn’t. For example, instead of investigating different age groups, one could compare nationalities. The most valuable results would come from testing two nationalities, one native English speaking and the other foreign language.
Despite my hypothesis having been supported, there were certain limitations to this experiment. Firstly, the sample of 14 and 15 year olds should have been equal in order to form valid comparisons between them. Furthermore, the idea of experimenter bias could cause anomalies in our data. This can be controlled either by having a single experimenter or by closely monitoring all contact between subject and experimenter. Another variable that influenced our results was the nationality and background of each participant; whether or not English is their first language.
Conclusion
In conclusion, the statistical data helps prove my hypothesis, that distinguishing colors would take longer than reading words. This is due to the fact that the participant is provided with conflicting stimuli for the former, which lengthens the time taken to respond.
WORD COUNT: 1990
References
"Attention." Wikipedia. 20 Mar. 2009 .
Engel-Andreasen, Michael. The Effect of Interfering Word Stimuli Upon Naming Colors Serially. Electronic.
"STUDIES OF INTERFERENCE IN SERIAL VERBAL REACTIONS." Journal of Experimental Psychology 18: 643-662. Abstract. Ed. Christopher D. Green. 31 Mar. 2009 .
William, Glassman E., and Hadad Marilyn, eds. Approaches to Psychology . 5th ed. Berkshire: Open University Press, 2009.
Appendices
Appendix 1: Calculations
Age 14: 20.20 – 9.88 = (10.32/20.20)*100 = 51% (Words took 51% less time than colors)
Age 15: 26.06 – 10.55 = (15.51/26.06)*100 = 60% (Words took 60% less time than colors)